1 Experimental study of GFRP-concrete hybrid beams with low degree of 1 shear connection 2 3 Catalin A. Neagoe a,b,* , Lluís Gil a,b , Marco A. Pérez b,c 4 a Department of Strength of Materials and Structural Engineering, Universitat Politècnica de Catalunya – 5 BarcelonaTech, Jordi Girona 31, 08034 Barcelona, Spain 6 b Laboratory for the Technological Innovation of Structures and Materials (LITEM), Colon 11, TR45, 7 Terrassa, 08222 Barcelona, Spain 8 c Institut Químic de Sarria, Universitat Ramon Llull, Via Augusta 390, 08017 Barcelona, Spain. 9 10 Abstract 11 Recent developments in the design of advanced composite materials for construction have led researchers to 12 create novel high-performance structural elements that combine fiber-reinforced polymer (FRP) shapes with 13 traditional materials. The current study analyzes the experimental structural response of eight hybrid beams 14 made of pultruded glass FRP (GFRP) profiles mechanically connected to reinforced concrete (RC) slabs, 15 suitable for building floors as well as footbridge and marine pier superstructures. The influence of partial 16 interaction is studied by considering a low degree of shear connection and an analytical assessment of the 17 whole response is carried out using previous formulations, highlighting a good accuracy. The behavior of the 18 hybrid beams is further evaluated against that of equivalent reinforced concrete beams and single GFRP 19 profiles, thus proving the feasibility of the solution. 20 * Corresponding author at: Laboratory for the Technological Innovation of Structures and Materials (LITEM), Colon 11, TR45, Terrassa, 08222 Barcelona, Spain. Tel.: +34 9373 98 727. E-mail address: [email protected] (C.A. Neagoe).
31
Embed
Experimental study of GFRP-concrete hybrid beams with low ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Experimental study of GFRP-concrete hybrid beams with low degree of 1
shear connection 2
3
Catalin A. Neagoe a,b,*, Lluís Gil a,b, Marco A. Pérez b,c 4
a Department of Strength of Materials and Structural Engineering, Universitat Politècnica de Catalunya – 5
interlaminar shear; (f) full section effective moduli. 151
The second component of the tested beams, the concrete, was produced in two different compositions using a 152
rapid hardening cement class 42.5 R, each mix corresponding to a batch of five beams. Average concrete 153
compressive strength was determined 28 days after fabrication on cubic samples, following EN 12390. The 154
remaining mechanical properties were obtained using the relations provided by Eurocode 2 [28] and are listed 155
together in Table 2. 156
Table 2. Mechanical properties of concrete mixes: average compressive strength (𝑓𝑓𝑐𝑐𝑐𝑐), modulus of elasticity 157 (𝐸𝐸𝑐𝑐) and average tensile strength (𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐). 158
Concrete mix Cement type and class 𝑓𝑓𝑐𝑐𝑐𝑐 (MPa) 𝐸𝐸𝑐𝑐 (GPa) 𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 (MPa) C1 CEM II/A – 42.5 R 30.0 28.6 1.90 C2 CEM II/A – 42.5 R 35.0 30.0 2.21
159
Steel reinforcement bars used in the fabrication of the hybrid beams were of class B500S, with a yielding 160
strength of 500 MPa and modulus of elasticity of 200 GPa. 161
8
162
2.3. Proposed hybrid designs and fabrication 163
Two different models of hybrid GFRP-concrete beams were designed to be tested and analyzed in the 164
investigation. Entitled M1 and M2, the hybrid models differed in the type of concrete cross-section geometry. 165
In addition to these two, an equivalent reinforced concrete model, designated M0, was included to serve as 166
reference in the analysis. All members had 2000 mm in length and 170 mm in height, with a top concrete slab 167
of 400x50 mm. Fig. 3 illustrates the constructive details of the specimens. 168
169
Fig. 3. Constructive details of the tested specimens: cross-section models with top view and side view of M1 170
and M2 hybrid beams (mm). 171
The hybrid beams were all made of a GFRP profile that was attached to the bottom of a concrete slab by 172
means of steel shear connectors. In contrast to model M2, model M1 had the profile also laterally encased in 173
concrete, forming a T-shaped composite member. The reinforced concrete model M0 featured a similar cross-174
section to M1 but instead of the GFRP profile the beam had an equivalent area of steel rebars capable of 175
9
producing a theoretically similar tensile force as the profile working under partial interaction conditions. Shear 176
connectors were installed before concreting of the hybrid beams, in pre-drilled holes located alternatively at 177
100 mm along the profile’s upper flange, as seen in Fig. 4. M6 steel bolts with a class resistance of 8.8 and 178
ultimate shear strength of 480 MPa were manually fastened into position with a torque of 10 Nm. The small 179
diameter of the shanks coupled with the longitudinal alternate distribution allowed for the desired 180
development of partial shear interaction. As a side note, for beams model M1 there was no lateral connection 181
between the GFRP profile and the concrete, and for beams model M2 the profile’s support regions were 182
encased in 200 mm wide concrete blocks (Fig. 3, top and side views) so as to prevent a premature local 183
crushing failure, as recommended by initial small-scale bending tests. 184
185
Fig. 4. Fabrication process for the hybrid beams: (a) installment of steel bolts; (b) completed formwork; (c) 186
concrete casting; (d) specimens prior to instrumentation and testing. 187
In order to maintain the integrity of the concrete slab during transportation and testing, 5Ø8 mm steel bars 188
were placed at its center as constructive longitudinal reinforcement. Transverse steel reinforcement was 189
provided only at the middle and at the ends of the slab. Because the investigation focused on the flexural 190
behavior of the beams, reference model M0 had only constructive transverse reinforcement in addition to the 191
3Ø12 bottom longitudinal bars. Reinforcement concrete cover was in all cases 20 mm. 192
Ten beams were fabricated using the three model designs: two units of model M0, four of model M1 and four 193
of model M2. They were subsequently divided in two groups of five specimens according to the 194
corresponding test setup. 195
196
10
2.4. Test setup and procedure 197
A month after fabrication the beams were subjected to positive moments in a three-point or four-point bending 198
test configuration. All specimens were simply supported over a span of 1800 mm, either on elastomeric pads 199
(test setup I) or on 40 mm wide steel cylinders (test setup II). A pair of Isolgomma 200x200x20 mm 200
elastomeric supports with a density of 0.7 kg/dm3 was used initially but, after the first batch of tests, results 201
showed that this measure was too conservative taking into account that the ends of the profiles were encased 202
in concrete. 203
In test setup I the beams were loaded at the midspan using a 250 kN MTS hydraulic actuator. A 45x20 mm 204
wood piece was used to spread the concentrated load from the actuator head to the top of the specimen. In 205
contrast, in test setup II the applied load produced by a 500 kN capable actuator was distributed in two parts 206
situated approximately at a third of the span by a steel frame with semi-cylindrical supports. Loading was 207
performed in a quasi-static manner with a constant displacement rate of 2 mm/min. Details of both test setups 208
are illustrated in Fig. 5. 209
210
Fig. 5. Schematic of load arrangements and instrumentation of hybrid beams (mm). 211
Instrumentation was similar for both configurations so as to record and compare similar parameters of the 212
flexural behavior. Deflections were measured at the midspan and at 500 mm towards each of the supports by 213
11
RIFTEK laser triangulation sensors. In the case of the beams placed on elastomeric pads the vertical 214
displacements of the supports were registered by two Waycon LRW-M-100-S linear potentiometers. The 215
hybrid specimens were additionally instrumented at one end with an HBM WA/20 displacement transducer 216
(LVDT) so as to record the relative slip between the top flange of the GFRP profile and concrete slab. 217
Strain gauges were attached in key sections of the beams, near or at the center span and at 150 mm from one 218
of the supports. For beams model M2 axial strains were measured across the concrete slab and the GFRP 219
profile, in sections S1 and S2 (Fig. 5). In this way the slip strain between the two constitutive materials could 220
be measured. In section S2 a couple of strain gauge rosettes were placed on the profile’s web to determine the 221
angular strains in the composite material. Hybrid beams model M1 were instrumented just in section S1 and 222
along the bottom flange of the profile. The control or reference specimens, represented by the M0 reinforced 223
concrete beams and the single GFRP profiles, were tested in similar configurations to those illustrated in Fig. 224
6. 225
226
Fig. 6. Laboratory setup and instrumentation: (a) Profile 2 test; (b) M2 hybrid beam in test setup I; (c) M2 227
hybrid beam in test setup II. 228
Data measured by the sensors were gathered by an HBM MGCplus data acquisitioning system at a rate of 50 229
Hz. In the case of the M2 hybrid beams a high speed camera was used to capture the development of the 230
brittle failure at a speed of 2000 fps. 231
The beams were split in two groups according to the loading scheme that was applied. The characteristics of 232
the two are outlined in Table 3 and the experimental results will be presented and discussed in the following 233
section. 234
12
Table 3. Characteristics of test specimens. 235
Beam ID Test setup Type Model Weight (kN/m)
Concrete mix
M0-RCB1 I RC M0 1.03 C1 M1-HB1 I GFRP-RC M1 1.02 C1 M1-HB2 I GFRP-RC M1 1.02 C2 M2-HB1 I GFRP-RC M2 0.61 C1 M2-HB2 I GFRP-RC M2 0.61 C2 Profile 1 I GFRP 0.03 Profile 2 a I GFRP 0.03 M0-RCB2 II RC M0 1.03 C2 M1-HB3 II GFRP-RC M1 1.02 C1 M1-HB4 II GFRP-RC M1 1.02 C2 M2-HB3 II GFRP-RC M2 0.61 C1 M2-HB4 II GFRP-RC M2 0.61 C2
a Had wood block stiffeners placed at the critical sections (reaction points). 236
237
3. Results and discussion 238
3.1. Experimental results 239
3.1.1. Flexural behavior and failure modes 240
In order to assess the performance of the different tested beams, the structural behavior of the control 241
specimens is discussed foremost. Beam M0-RCB1 had the typical ductile flexural response of a reinforced 242
concrete element, failing by yielding of the bottom steel reinforcing bars accompanied by crushing of the 243
concrete top. In the case of M0-RCB2 tested under four-point bending, the failure was sudden due to a 244
diagonal tensile shear crack formed near one of the supports, justified by the low ratio of transverse 245
reinforcement. GFRP Profile 1 and 2 failed both due to a loss of global stability, by lateral torsional buckling, 246
as illustrated in Fig. 7. Consequently, after the initial failure of Profile 2, the uneven normal compressive 247
stress provoked a local buckling of the top flange observable in the same figure. The flexural behavior of the 248
two specimens was linear elastic, with Profile 2 attaining a higher capacity because of the stiffeners placed at 249
the reaction points. Profile’s 2 ultimate capacity was just 17% lower than that of M0-RCB1, with a maximum 250
deflection 3 times as great. 251
13
252
Fig. 7. Buckling failure modes of profile control specimens: (a) Profile 1; and (b) Profile 2. 253
Hybrid beams M1-HB1 and M1-HB2, which were made of a GFRP structural profile encased in a T-shaped 254
concrete beam, displayed a generally bilinear response up to ~90% of the ultimate load, a superior strength in 255
comparison to M0-RCB1 and double the flexural rigidity of the single profiles. Furthermore, the maximum 256
load sustained by M1-HB2 represented a threefold increase over the value recorded for Profile 1. The bilinear 257
shape of the responses is attributed mainly to the change in the stress transfer mechanism at the connection 258
level. Thus, the initial slope reflects a complete interaction between the two layers while the second a partial 259
interaction (i.e., flexible connection). At the beginning of the tests, large vertical flexural cracks appeared in 260
the concrete web of the hybrid beams due to the material’s loss of tensile strength, as revealed by the jumps in 261
the load-displacement responses represented in Fig. 8. As loading continued, the cracks progressed towards 262
the inferior central part of the top slab. Failure of the M1 hybrid elements began with crushing of the concrete 263
top at the midspan and ended a few moments later when the profile’s bottom flange suddenly detached from 264
the GFRP web. The cause of the brittle collapse was determined to be the increased shear stress which had 265
developed at the web-flange junctions, at the ends of the pultruded members. After failure, the two hybrid 266
beams continued to work in flexure, displaying a recovery capacity of up to 75% of the maximum sustained 267
load. 268
14
269
Fig. 8. Experimental bending results under test setup I: load-midspan deflection curves until failure. 270
The flexural responses of hybrid beams M2-HB1 and M2-HB2, which were made of a GFRP structural profile 271
attached with steel bolts to a reinforced concrete slab, were similar to those of the previous M1 hybrid beams. 272
Slight differences are visible in Fig. 8 in the increased deformability explained by the fact that the GFRP web 273
was not laterally encased in concrete and in the higher nonlinear response towards the end, justified by the 274
concrete’s constitutive behavior under high compressive strains. This time around the flexural cracks were 275
less wide and more spread across the slab, starting especially from the connectors’ positions and reaching 276
towards the edges and central line. For M2-HB1 failure began with crushing of the concrete top followed by a 277
brittle shear delamination at one of its ends, at the junction between the GFRP profile’s top flange and web. 278
The shear failure dispersed immediately towards the midspan of the beam, causing an additional vertical 279
displacement of the steel bolts and a local buckling of the compressed web (post-failure mechanism). In 280
contrast, failure of hybrid beam M2-HB2 occurred suddenly at the midspan, without concrete crushing, in the 281
zone directly placed under the applied load, possibly being induced by a fracture of the load spreading piece. 282
Thus, a high compressive stress present at the top of the GFRP profile determined a crushing type of failure to 283
occur in the profile’s web followed by longitudinal delaminations of the composite material. No significant 284
recovery capacity was displayed by the M2 hybrid beams during the three-point bending tests. 285
In the second phase of the experimental campaign, the hybrid specimens tested under four-point bending 286
exhibited a generally bilinear structural response and a higher capacity than control beam M0-RCB2. 287
Nevertheless, the flexural stiffness was lower, with M2-HB3 and M2-HB4 experiencing a greater 288
15
deformability as seen in Fig. 9. The occurrence of flexural cracks is emphasized again by the sudden drops in 289
load-bearing capacity, especially in the initial stage for the M1 beams. Opposite to the first testing phase, all 290
hybrid beams failed in the same manner due to a longitudinal shear crack which developed at the top web-291
flange junction, without any prior crushing of the reinforced concrete slab. M1-HB3 and M1-HB4 retained 292
after failure a capacity of 50-60% of the maximum load whilst beams model M2 provided inconclusive 293
recovery results. 294
295
Fig. 9. Experimental bending results under test setup II: load-midspan deflection curves until failure. 296
Fig. 10 illustrates the main failure modes observed for the hybrid beams during the experimental campaign. 297
Cracks generated by the inward slip of the profile were marked in red in Fig. 10c. 298
299
Fig. 10. Failure modes of hybrid beams: (a) profile web-flange shear preceded by crushing of the concrete 300
slab; (b) crushing of the profile’s web; (c) profile web-flange shear. 301
Load-deflection charts reflect also the change in the slab’s compressive strength, whereas beams fabricated 302
using concrete mix C2 have a slightly higher flexural stiffness and capacity, as expected. In spite of this, the 303
16
ultimate load of the hybrid beams seems to be limited by the amount of bending deformation supported and 304
more precisely by the amount of shear that the GFRP profile can carry. Considering a uniform distribution of 305
the shear stress in the web of the profile and neglecting the contribution of the flanges, Fig. 11 plots the 306
variation of the shear force percentage carried by the composite profile against the applied load. 307
308
Fig. 11. Hybrid beam M2-HB4, section S2: shear force carried by the profile’s web in function of the applied 309
load. 310
311
3.1.2. Composite action and interlayer slip 312
Gauge measurements performed at sections S1 and S2 were used to plot the variation of the longitudinal 313
strains in function of the applied load. Fig. 12 illustrates this variation for the particular case of hybrid beam 314
M2-HB4. Similar strains across the top slab suggest that the whole width of the concrete section was effective. 315
This result is in agreement with the design code recommendations for simply supported steel-concrete 316
composite beams [29]. Negative strain values registered on the top flange of the GFRP profile indicate that the 317
pultruded element started to work in compression at higher load levels. For the specimens which failed 318
primarily due to slab crushing, concrete strain curves displayed maximum negative values in the vicinity of 319
0.35%. Maximum GFRP axial deformations were in the range of 1.1% for the beams tested under three-point 320
bending, respectively 0.6% for the specimens under four-point bending. 321
17
322
Fig. 12. Hybrid beam M2-HB4, section S1: variation of axial strains in function of the applied load. 323
The same data was used to plot the axial strains as a function of the beam’s depth for different load levels. In 324
this way, a better view of the composite action developing in the hybrid beam was obtained, exemplified here 325
by Fig. 13 for hybrid beam M2-HB4. After 20 kN of load there was an increased slip strain developed 326
between the concrete slab and the profile that led to the appearance of two neutral axes in the cross-section of 327
the element. The first neutral axis of the T-shaped beam laid in the top concrete slab, close to the steel 328
reinforcement level, while the position of the second neutral axis moved from the connection level towards the 329
center of the composite member. Due to the relatively low elastic modulus of GFRP, shear has an important 330
role in the behavior of short elements (height/span < 1/20) in the sense that at high stress levels the section 331
does not remain plane after bending. This warping effect of the profile is slightly noticeable in Fig. 13. 332
333
Fig. 13. Hybrid beam M2-HB4, section S1: normal strain distribution at different load levels (kN). 334
18
Axial strain variations registered for the M2 specimens at section S2, near one of the supports, also point out 335
that the web was in compression at higher loads; however, the top flange was still submitted to tensile 336
deformations, being hindered by the mechanical connection. It is believed that this effect coupled with the 337
significant in-plane shear deformation of the profile led to the web-flange shear failure of the hybrid beams. 338
The relative slip between the profile and the slab at the end of the hybrid beams is plotted in Fig. 14 against 339
the load ratio. Hybrid beams model M1 presented a complete shear interaction up to 40% of their ultimate 340
flexural capacity whereas beams model M2 had a weaker shear interaction starting from about 25%. The 341
average maximum slip was 1.7 mm for type M1 and an almost double amount of 3.5 mm for type M2. 342
Overall, hybrid beams model M1 displayed a stiffer, higher composite action due to the concrete web which 343
prevented the steel bolts and GFRP profile form sliding too much. In addition, the concrete class had a similar 344
influence, with higher strengths limiting the slip to a greater degree. 345
346
Fig. 14. Relative end slip of the profile versus load ratio. 347
The slip strain-bending moment curves plotted in Fig. 15 illustrate similar nonlinear responses for the M2 348
specimens. The exception resides in the fact that during the first testing phase the deformations attained were 349
double in comparison with the results from the second testing phase, under four-point bending. 350
19
351
Fig. 15. M2 hybrid beams: slip strain variation in function of the applied bending moment. 352
The partial interaction effects attributed to the low degree of shear connection and flexibility of the bolts were 353
noticed not only from numerical data but also during a visual inspection of the tested members, as illustrated 354
in Fig. 16. 355
356
Fig. 16. Visual evidence of partial interaction: (a) occurrence of slip; (b) deformation of bolts; (c) distortion of 357
connector holes. 358
359
3.2. Analytical assessment 360
Based on the analytical procedure for PFRP-RC hybrid beams detailed by the present authors in [30], the data 361
obtained during the experimental campaign were compared with theoretical results. The procedure used is 362
built on the Timoshenko beam theory and on the elastic interlayer slip model, where partial interaction effects 363
over flexural capacity, deflection and strains are quantified by using a dimensionless parameter which relies 364
mainly on the connection’s stiffness and beam rigidity characteristics. The shear resistance of the hybrid 365
20
beams is considered to be entirely provided by the GFRP profile, which represents a conservative but reliable 366
approach as suggested by the same study. For the sake of completeness, Appendix A summarizes the main 367
mathematical relations used herein. 368
A comparison is made in Table 4 between the experimental and analytical results, at the serviceability and 369
ultimate limit states (SLS and ULS). The numbers reveal a difference under 5% between the ultimate bending 370
moments and an even smaller difference for the serviceability case. There are two larger exceptions for SLS 371
because of the jumps in deflection measurements induced by the occurrence of large flexural cracks. The 372
differences between maximum deflections at failure are underestimated because the concrete behavior is 373
considered linear in the analytical model. Nevertheless, the overall stiffness of the hybrid beams considering 374
partial interaction is estimated with good precision as it will be shown and discussed later. The ratio between 375
the bending moment considering a shear type of failure and the one based on a compressive failure of the 376
concrete slab, 𝑀𝑀𝑢𝑢/𝑀𝑀𝑢𝑢𝑐𝑐𝑐𝑐, reveals an important ductility aspect of hybrid beams. For the four-point bending test 377
configuration the ratios are less than unity while for the three-point setup most of the values are slightly over 378
it. This theoretical evaluation coincides with the experimental observations, where three of the eight hybrid 379
beams, M1-HB1, M1-HB2 and M2-HB1, failed in a pseudo-ductile manner while on the contrary the rest had 380
a predominantly fragile response. 381
382
Table 4. Results for hybrid beams at the serviceability (SLS) and ultimate limit states (ULS/u): bending 383
moment (𝑀𝑀), midspan deflection (𝑤𝑤), and bottom flange ultimate axial stress (𝜎𝜎). 384
Beam Experimental Analytical Failure mode 𝑀𝑀𝑆𝑆𝑆𝑆𝑆𝑆 a
for building floors as well as footbridge and marine pier superstructures. Because the flexural behavior of a 439
hybrid element relies greatly on the connection system, a low degree of shear interaction was considered in 440
this work to study its effects. Lastly, a comparative analysis between experimental and analytical results was 441
carried out. The following conclusions are drawn from the investigation: 442
• Glass FRP-concrete hybrid beams prove to be structurally-efficient elements with a high flexural 443
capacity to self-weight ratio. 444
• Cost-effective solutions can be obtained by using off-the-shelf materials and connectors as opposed to 445
custom-made composite systems. 446
25
• Compared to the single pultruded GFRP profiles, the hybrid beams had superior bending resistance, 447
pseudo-ductile behavior in specific cases and no instability type of failure. Furthermore, the 448
composite material was better used, being subjected to a stress up to 80% of its tensile strength. 449
• Compared to the equivalent reinforced concrete beams, the hybrid specimens displayed ~50% higher 450
ultimate capacity with 50% less weight. Nevertheless, the flexural stiffness was lower due to the 451
elastic modulus of the GFRP and more substantially due to the low degree of partial interaction. 452
• Two types of cross-section hybrid models – M1 and M2 – were considered in the experimental 453
campaign, the difference residing in the lateral confinement of the profile for the first model. Overall, 454
similar responses were registered for both types; however, beams M1 had a more rigid mechanical 455
connection with slip values at half of those of M2, a more linear load-midspan deflection response and 456
at least a 50% recovery capacity after collapse. Nonetheless, the M2 beams are more suited from a 457
practical point of view due to the reduced self-weight. A stronger interlayer connection would 458
compensate for the lack of lateral confinement. 459
• Normal strength concrete allowed for a pseudo-ductile type of failure, where crushing of the concrete 460
slab constituted a warning sign of the imminent collapse. 461
• The increase in concrete strength improved the ultimate bending capacity and stiffness but the 462
behavior was still limited by the shear deformation that the profile could bear. 463
• Rupture of the GFRP profile’s web-flange junction constituted the primary type of failure and was 464
produced mainly by high shear stress concentrations. As observed, the web-flange junction represents 465
a transition area where the internal microstructure of the composite shape changes drastically and 466
where the mid-plane multidirectional transverse reinforcement ends for the profiles used in the 467
investigation. 468
• A transverse crushing failure of one of the specimens indicates that stiffeners should be placed under 469
areas with concentrated loads to prevent a premature type of collapse. Profiles with hollow sections 470
could also defer the occurrence of brittle failure modes as well as increase the flexural capacity. 471
26
• The analytical results matched the experimental results with good precision. Theoretical flexural 472
responses highlighted a positive agreement in terms of bending moments, deflections and normal 473
stresses. Furthermore, the analytical model was able to capture the influence of the partial interaction 474
on these values. 475
476
Acknowledgements 477
The presented work is part of research project COMPOBEAM – Researching the flexural behavior of mixed 478
beams made from concrete and GFRP profiles, developed at CER LITEM/UPC-BarcelonaTech. The authors 479
would like to acknowledge the financial support from PIGRA Engineering S.L. through CDTI. The first 480
author is also grateful for the financial aid provided by the FPI-UPC doctoral scholarship. 481
482
Appendix A. Analytical formulations 483
A brief description of the main mathematical expressions discussed by the authors in [30] and used in the 484
current analytical assessment is adjoined. 485
The maximum deflection of a PFRP-RC hybrid beam under flexure considering a complete shear interaction 486
behavior is expressed as a sum of the deflection due to bending and shear deformation: 487
𝑤𝑤𝑐𝑐𝑐𝑐 =𝑓𝑓
𝐸𝐸𝐼𝐼𝑐𝑐𝑐𝑐+
𝑔𝑔𝐺𝐺𝑝𝑝𝐴𝐴𝑤𝑤
(A.1)
where 𝑓𝑓 and 𝑔𝑔 are functions given by the elasticity theory which depend on the load and supporting 488
conditions; 𝐸𝐸𝐼𝐼𝑐𝑐𝑐𝑐 is the flexural rigidity of the member; 𝐺𝐺𝑝𝑝 the shear modulus of the profile and 𝐴𝐴𝑤𝑤 the 489
profile’s web area. 490
Under partial shear interaction conditions the maximum deflection of a hybrid beam is approximated as: 491
𝑤𝑤𝑝𝑝𝑚𝑚 = (1 + 𝜉𝜉)𝑓𝑓
𝐸𝐸𝐼𝐼𝑐𝑐𝑐𝑐+
𝑔𝑔𝐺𝐺𝑝𝑝𝐴𝐴𝑤𝑤
(A.2)
where 𝜉𝜉 represents a dimensionless partial interaction parameter defined as: 492
27
𝜉𝜉 = �𝐸𝐸𝐼𝐼𝑐𝑐𝑐𝑐𝐸𝐸𝐼𝐼0
− 1� �1 + �𝛼𝛼𝛼𝛼𝜋𝜋�2�−1
(A.3)
The composite action parameter 𝛼𝛼𝛼𝛼 is computed using the following relationship: 493
𝛼𝛼𝛼𝛼 = �𝐾𝐾𝑐𝑐𝑑𝑑𝑐𝑐2
𝑠𝑠𝑐𝑐𝐸𝐸𝐼𝐼𝑐𝑐𝑐𝑐
𝐸𝐸𝐼𝐼0(𝐸𝐸𝐼𝐼𝑐𝑐𝑐𝑐 − 𝐸𝐸𝐼𝐼0)𝛼𝛼 (A.4)
where 𝛼𝛼 represents the beam’s span; 𝐸𝐸𝐼𝐼0 the flexural rigidity of the hybrid member under no shear interaction; 494
𝐾𝐾𝑐𝑐 and 𝑠𝑠𝑐𝑐 the connector’s stiffness and spacing (pitch), respectively; and 𝑑𝑑𝑐𝑐 the distance between the centroids 495
of the slab and profile. 496
The flexural capacity, 𝑀𝑀, of a hybrid element is found from the equilibrium of the cross-section. For the 497
partial interaction case the slip strain must also be determined. An approximate solution that excludes 498
calculating the slip strain, denoted 𝑀𝑀𝑒𝑒𝑒𝑒𝑒𝑒, is based on the partial interaction parameter 𝜉𝜉 and is expressed as: 499
𝑀𝑀𝑒𝑒𝑒𝑒𝑒𝑒 = 𝑀𝑀 �1 − 𝜉𝜉ℎ𝑝𝑝𝐸𝐸𝑝𝑝6𝐸𝐸𝐼𝐼𝑐𝑐𝑐𝑐
�ℎ𝑐𝑐𝐴𝐴𝑝𝑝 + ℎ𝑝𝑝𝐴𝐴𝑤𝑤�� (A.5)
where ℎ𝑝𝑝 is the height of the profile; 𝐸𝐸𝑝𝑝 the profile’s flexural modulus; ℎ𝑐𝑐 the height of the concrete slab; and 500
𝐴𝐴𝑝𝑝 the profile’s transverse area. 501
The maximum axial stress evaluated at the bottom of the member is obtained from: 502
𝜎𝜎𝑐𝑐𝑚𝑚𝑚𝑚 = 𝑀𝑀��1 −𝐸𝐸𝐼𝐼0𝐸𝐸𝐼𝐼𝑐𝑐𝑐𝑐
(1 + 𝜉𝜉)�1
𝐴𝐴𝑝𝑝𝑑𝑑𝑐𝑐+
0.5𝐸𝐸𝑝𝑝ℎ𝑝𝑝𝐸𝐸𝐼𝐼𝑐𝑐𝑐𝑐
(1 + 𝜉𝜉)� (A.6)
The structural capacity of a hybrid beam is also limited by the amount of shear force that the profile can carry, 503
which is computed from: 504
𝑉𝑉𝑐𝑐𝑚𝑚𝑚𝑚 = 𝜏𝜏𝑐𝑐𝑚𝑚𝑚𝑚𝐴𝐴𝑠𝑠ℎ (A.7) where 𝜏𝜏𝑐𝑐𝑚𝑚𝑚𝑚 is the in-plane shear strength of the composite material and 𝐴𝐴𝑠𝑠ℎ the sheared area of the profile, 505
typically considered as the area of the web. 506
507
References 508
[1] Bansal A, Monsalve Cano JF, Osorio Muñoz BO, Paulotto C. Examples of Applications of Fibre 509 Reinforced Plastic Materials in Infrastructure in Spain. Struct Eng Int 2010;20:414–7. 510 doi:10.2749/101686610793557726. 511
28
[2] Areiza Hurtado M, Bansal A, Paulotto C, Primi S. FRP girder bridges: Lessons learned in Spain in the 512 last decade. 6th Int. Conf. FRP Compos. Civ. Eng. (CICE 2012), Rome, Italy: 2012. 513
[3] Cheng L, Karbhari VM. New bridge systems using FRP composites and concrete: A state-of-the-art 514 review. Prog Struct Eng Mater 2006;8:143–54. doi:10.1002/pse.221. 515
[4] Chen D, El-Hacha R. Hybrid FRP-Concrete Structural Member: Research and Development in North 516 America. In: Ye L, Feng P, Yue Q, editors. 5th Int. Conf. FRP Compos. Civ. Eng. (CICE 2010), 517 Beijing, China: Springer & Tsinghua University Press; 2011, p. 185–90. 518
[5] Chen D, El-Hacha R. Hybrid FRP-Concrete Structural Member: Research and Development in Europe 519 and Asia. In: Ye L, Feng P, Yue Q, editors. 5th Int. Conf. FRP Compos. Civ. Eng. (CICE 2010), 520 Beijing, China: Springer & Tsinghua University Press; 2011, p. 191–6. 521
[6] Keller T. Recent all-composite and hybrid fibre-reinforced polymer bridges and buildings. Prog Struct 522 Eng Mater 2001;3:132–40. doi:10.1002/pse.66. 523
[7] Clarke JL, editor. Structural Design of Polymer Composites - EUROCOMP Design Code and 524 Handbook. E & FN SPON; 1996. 525
[8] Nordin H, Täljsten B. Testing of hybrid FRP composite beams in bending. Compos Part B Eng 526 2004;35:27–33. doi:10.1016/j.compositesb.2003.08.010. 527
[9] Correia JR, Branco FA, Ferreira J. GFRP-concrete hybrid cross-sections for floors of buildings. Eng 528 Struct 2009;31:1331–43. doi:10.1016/j.engstruct.2008.04.021. 529
[10] Neagoe CA, Gil L. Evaluation of Deflections for PFRP-RC Hybrid Beams with Complete and Partial 530 Shear Connection. In: Chiorean CG, editor. Proc. Second Int. Conf. PhD Students Civ. Eng. Archit. 531 (CE-PhD 2014), Cluj-Napoca, Romania: UTPRESS; 2014, p. 57–64. 532
[11] Saiidi M, Gordaninejad F, Wehbe N. Behavior of Graphite/Epoxy Concrete Composite Beams. J Struct 533 Eng 1994;120:2958–76. doi:10.1061/(ASCE)0733-9445(1994)120:10(2958). 534
[12] Sekijima K, Miyata K, Ihara T, Hayashi K. Study on Flexural Behavior of Fiber Reinforced Plastic-535 Concrete Composite Beam. Third Int. Symp. Non-Metallic Reinf. Concr. Struct. (FRPRCS-3), 536 Sapporo, Japan: Japan Concrete Institute; 1997, p. 543–50. 537
[13] Sekijima K, Ogisako E, Miyata K, Hayashi K. Analytical study on flexural behavior of GFRP-concrete 538 composite beam. In: Teng JG, editor. Int. Conf. FRP Compos. Civ. Eng. (CICE2001), vol. II, Hong 539 Kong, China: Elsevier Science Ltd.; 2001, p. 1363–70. 540
[14] Biddah A. Experimental investigation of pultruded FRP section combined with concrete slab. In: Tan 541 KH, editor. Sixth Int. Symp. FRP Reinf. Concr. Struct. (FRPRCS-6), Singapore: World Scientific 542 Publishing Co. Pte. Ltd.; 2003. 543
[15] Fam A, Skutezky T. Composite T-Beams Using Reduced-Scale Rectangular FRP Tubes and Concrete 544 Slabs. J Compos Constr 2006;10:172–81. doi:10.1061/(ASCE)1090-0268(2006)10:2(172). 545
[16] Hai ND, Mutsuyoshi H, Asamoto S, Matsui T. Structural behavior of hybrid FRP composite I-beam. 546 Constr Build Mater 2010;24:956–69. doi:10.1016/j.conbuildmat.2009.11.022. 547
[17] Mendes Ferreira AJ, Ribeiro MCS, Torres Marques A. Analysis of hybrid beams composed of GFRP 548 profiles and polymer concrete. Int J Mech Mater Des 2004;1:143–55. 549
29
[18] Mendes PJD, Barros JAO, Sena-Cruz JM, Taheri M. Development of a pedestrian bridge with GFRP 550 profiles and fiber reinforced self-compacting concrete deck. Compos Struct 2011;93:2969–82. 551 doi:10.1016/j.compstruct.2011.05.005. 552
[19] Mutsuyoshi H, Shiroki K, Hai N, Ishihama T. Composite Behavior of a Pultruded Hybrid CFRP-GFRP 553 Beam with UFC Deck. In: Ye L, Feng P, Yue Q, editors. 5th Int. Conf. FRP Compos. Civ. Eng. (CICE 554 2010), Beijing, China: Springer; 2011, p. 111–4. 555
[20] Chen D, El-Hacha R. Damage tolerance and residual strength of hybrid FRP–UHPC beam. Eng Struct 556 2013;49:275–83. doi:10.1016/j.engstruct.2012.11.016. 557
[21] Deskovic N, Triantafillou TC, Meier U. Innovative Design of FRP Combined with Concrete: Short-558 Term Behavior. J Struct Eng 1995;121:1069–78. doi:10.1061/(ASCE)0733-9445(1995)121:7(1069). 559
[22] Chakrabortty A, Khennane A, Kayali O, Morozov E. Performance of outside filament-wound hybrid 560 FRP-concrete beams. Compos Part B Eng 2011;42:907–15. doi:10.1016/j.compositesb.2011.01.003. 561
[23] Mutsuyoshi H, Hai ND, Aravinthan T, Manalo A. Experimental investigation of HFRP composite 562 beams. 10th Int. Symp. Fiber Reinf. Polym. Reinf. Concr. Struct. (FRPRCS-10), Tampa, FL, United 563 States: 2011. 564
[24] El-Hacha R, Chen D. Behaviour of hybrid FRP–UHPC beams subjected to static flexural loading. 565 Compos Part B Eng 2012;43:582–93. doi:10.1016/j.compositesb.2011.07.004. 566
[25] Gonilha JA, Correia JR, Branco FA. Structural behaviour of a GFRP-concrete hybrid footbridge 567 prototype: Experimental tests and numerical and analytical simulations. Eng Struct 2014;60:11–22. 568 doi:10.1016/j.engstruct.2013.12.018. 569
[26] Neagoe CA. Design and analysis of PFRP and concrete hybrid beams. In: Pérez MA, editor. Advanced 570 applications of composite materials in civil works and buildings, OmniaScience Monographs; 2014, p. 571 205–36. doi:10.3926/oms.207 [in Spanish]. 572
[27] European Committee for Standardization (CEN). EN 13706. Reinforced plastics composites – 573 Specifications for pultruded profiles - Part 3: Specific requirements. Brussels, Belgium: 2002. 574
[28] European Committee for Standardization (CEN). EN 1992-1-1:2004. Eurocode 2: Design of concrete 575 structures – Part 1-1: General rules and rules for buildings. Brussels, Belgium: 2004. 576
[29] European Committee for Standardization (CEN). EN 1994-1-1:2004. Eurocode 4: Design of composite 577 steel and concrete structures – Part 1-1: General rules and rules for buildings. Brussels, Belgium: 2004. 578
[30] Neagoe CA, Gil L. Analytical procedure for the design of PFRP-RC hybrid beams including shear 579 interaction effects. Compos Struct 2015;132:122–35. doi:10.1016/j.compstruct.2015.04.054. 580
[31] European Committee for Standardization (CEN). EN 1995-1-1:2004. Eurocode 5: Design of timber 581 structures – Part 1-1: General – Common rules and rules for buildings. Brussels, Belgium: 2004. 582
[32] Gay D, Hoa S, Tsai S. Composite materials: design and applications. CRC Press; 2003. 583